Full text of DOI:10.1207/S15324796ABM2303_3

Appraised Control, Coping, and Stress in a Community Sample:
A Test of the Goodness-of-Fit Hypothesis
Sandra G. Zakowski, Ph.D.
Finch University of Health Sciences/The Chicago Medical School
Martica H. Hall, Ph.D.
Western Psychiatric Institute and Clinic, University of Pittsburgh Medical Center
Laura Cousino Klein, Ph.D.
Pennsylvania State University
Andrew Baum, Ph.D.
University of Pittsburgh Medical Center
ABSTRACT
INTRODUCTION
Lazarus and Folkman proposed one of the most compre-
hensive theories of stress and coping in the psychology litera-
ture, but many of their postulates have received little empirical
attention, and some of the existing research has yielded contra-
dictory findings. This longitudinal study sought to clarify the
associations among control appraisal, coping, and stress
within this theoretical framework. The theory postulates that
coping strategies used tend to match the level of appraised
controllability of the stressor (matching hypothesis). It further
states that the effects of problem-focused versus emotion-fo-
cused coping are moderated by the appraised controllability of
the stressor (goodness-of-fit hypothesis). An alternative to the
latter is the main-effects hypothesis, which states that prob-
lem-focused coping is generally more effective in reducing dis-
tress regardless of appraisal. These hypotheses were tested on
72 adults who completed questionnaires on coping and control
appraisal. Stress was assessed using self-report (Symptom
Checklist-90–Revised) and a behavioral measure (proofread-
ing task) at two times approximately 2 months apart. Ap-
praised control significantly predicted type of coping such that
greater control was associated with more problem-focused and
less emotion-focused coping. Although the main-effects hy-
pothesis was not supported, the goodness-of-fit hypothesis was
partly confirmed by a significant control by emotion-focused
coping interaction predicting both self-report and behavioral
measures of stress.
Stress has been conceptualized in various ways, but the most
comprehensive theoretical framework proposed to date is the
transactional model put forth by Lazarus and Folkman (1). This
model is built on the assumption that stress depends on a number
of subjective cognitive judgments that arise from the dynamic in-
terplay between person and environment. Stress can be defined as
a person–situation interaction which the individual appraises as
relevant to his or her well being and as taxing his or her coping re-
sources (2). No event or situation is considered to be inherently
stressful. Rather, the individual’s subjective judgment of the situ-
ation as threatening or harmful is what defines a stressor. Ap-
praisals also address judgments of the resources available to the
individual, such as coping strategies and the degree of control he
or she perceives to have in meeting the demands of the situation
(e.g., 3). These secondary appraisal processes guide the use of
specific coping strategies, and the effectiveness of these strategies
determines the reappraisal of the situation and the individual’s
psychological adjustment. This study addresses two basic hy-
potheses from the transactional model that have received little
empirical attention. First, we examine whether control appraisals
of life events guide individuals’ choices of coping strategies
(matching hypothesis). Second, the goodness-of-fit hypothesis
stating that the effectiveness of problem- versus emotion-focused
coping is moderated by control appraisals of the stressful event
(goodness-of-fit hypothesis) is addressed. This is be compared to
the alternative main-effects hypothesis, which postulates that
problem-focused strategies are inherently more effective irre-
spective of control appraisal.
(Ann Behav Med
2001, 23(3):158–165)
Coping may be defined as “constantly changing cognitive
and behavioral efforts to manage specific external and internal
demands that are appraised as taxing or exceeding the resources
of the person” (1, p. 141). In keeping with the dynamic nature of
the model, coping is not considered a style or personality trait
that remains stable across a variety of situations (e.g., 4) but is a
set of coping strategies that are available to be implemented to
match specific situations. Coping may take one of two general
forms: emotion-focused or problem-focused coping (1). Emo-
tion-focused coping strategies are directed toward altering the
individual’s emotional response to the stressor and include strat-
egies such as self-blame, wishful thinking, and avoidance. They
We acknowledge Laura Redwine, Ph.D. for her contributions to the re-
search project; Michele Hayward, M.S. and Janel Alberts, M.S. for
their help with data processing; and Kimberly Laubmeier, M.S. for her
comments on an earlier version of the article.
Reprint Address: S. G. Zakowski, Ph.D., Department of Psychology,
Finch University of Health Sciences/The Chicago Medical School,
3333 Green Bay Road, North Chicago, IL 60064. E-mail:
Zakowsks@finchcms.edu
© 2001 by The Society of Behavioral Medicine.
158

Volume 23, Number 3, 2001
Coping and Stress
159
are focused on internal emotional states, not on the situation that
triggered the emotional states. Problem-focused coping, by con-
trast, functions to alter the problem itself by directly acting on
the situation. Some coping strategies may serve both functions
simultaneously, such as seeking social support (e.g., 5).
coping by control interaction for major life events. However, the
goodness-of-fit hypothesis was only supported for a ratio of
problem- to emotion-focused coping but not for raw coping
scores. Vitaliano, DeWolfe, Maiuro, Russo, and Katon (13)
found that the percentage of problem-focused coping was nega-
tively correlated with depression regardless of appraised
changeability of the stressors; however, in support of the good-
ness-of-fit hypothesis, this correlation was significantly stron-
ger for changeable stressors. In addition, there was a trend (p <
.1) for a positive correlation between percent emotion-focused
coping and depression for situations appraised as changeable.
Conway and Terry (11) reported a significant control by
self-denigration (an emotion-focused coping strategy) interac-
tion effect on depression as predicted by the goodness-of-fit hy-
pothesis. It should be noted that at least two studies did not find
support for this hypothesis (6,12). The discrepant findings may
be attributed to use of different coping and distress measures,
different scoring methods of coping subscales, and different
study populations used.
The existing literature provides relatively little evidence for
some of the basic hypotheses of Lazarus and Folkman’s model
(1). This may be due to a general lack of empirical research di-
rectly addressing these hypotheses as well as some of the method-
ological inconsistencies and shortcomings of existing research.
Some of these issues are addressed in this study. For example, all
but one of the studies (12) reviewed used cross-sectional designs
that did not control for prior levels of distress. It could be argued
that, rather than the fit between appraisal and coping determining
distress, it may be prior levels of distress that determine both
choice of coping and subsequent levels of distress. Previous re-
search has also limited assessments of stress to self-report mea-
sures. Conclusions would be strengthened by a multimeasure as-
sessment approach and would reduce concerns of reporting
biases that may occur when using self-report only (e.g., 14,15).
This study, therefore, included a behavioral measure of stress—a
proofreading task (16), which has been used extensively in previ-
ous studies of stress and has been shown to be associated with
other measures of stress including self-report and physiological
measures (e.g., 14,17). Finally, due to controversy in the literature
regarding the relative validity of two different scoring methods of
the coping scales, this study tested all hypotheses using the raw
and relative scoring methods (5).
Decisions regarding the type of coping strategies used
should depend on appraisal of the situation. In most situations,
people will use both types of coping to varying degrees (2).
However, it has also been suggested that if the stressor is ap-
praised as controllable, the individual may be more likely to use
problem-focused and less likely to use emotion-focused coping
strategies than if the event is appraised as uncontrollable (1).
There is some empirical evidence to support this claim. For ex-
ample, controllable daily hassles have been associated with
higher levels of problem-focused coping in college students (6).
In addition, Folkman and Lazarus (7) reported work stressors
(appraised as more controllable) to be associated with increased
problem-focused coping and health-related stressors to be asso-
ciated with increased emotion-focused coping in a middle-aged
community sample. Finally, a study of uncontrollable problems
after a nuclear accident suggested that emotion-focused coping
was associated with better adjustment (8).
Some theories of defense mechanisms are based on the
premise that some mechanisms are inherently more adaptive
than others (e.g., 9). This theoretical viewpoint has sometimes
also been applied to coping theories such that problem-focused
strategies are often assumed to be superior to emotion-focused
strategies in promoting adjustment, regardless of the situation.
This main-effects hypothesis has received some limited empiri-
cal support in cross-sectional studies (10,11) and one longitudi-
nal study (12). However, these studies examined only isolated
coping strategies (11,12), and contradictory evidence has also
been reported (e.g., 6,10).
The discrepant findings cast doubt on the idea that prob-
lem-focused coping is universally more effective. It is likely
that, as Folkman et al. suggested, neither of these strategies is in-
herently adaptive or maladaptive (3) but that coping may have to
be considered as it interacts with the appraisal of the situation to
more reliably predict psychological adjustment. According to
the goodness-of-fit hypothesis, the effectiveness of a coping
strategy in reducing distress depends on the degree to which it
matches the situation as it is appraised. Control appears to be
particularly important in this regard. Controllable stressors may
be best dealt with by using coping that is focused on the problem
itself, whereas such efforts may be ineffective or harmful in the
face of an uncontrollable stressor about which little can be done
(8). Conversely, in the case of an uncontrollable problem, a
more emotion-focused approach may best reduce distress be-
cause one’s internal state may be more amenable to change than
the situation itself. Put simply, a “good fit” between appraisal
and coping would consist of the greater use of emotion-focused
coping when dealing with an uncontrollable stressor and greater
problem-focused coping for a controllable stressor.
This study tested the three basic hypotheses describing cop-
ing and control addressing some of the methodological prob-
lems of previous research. First, we examined whether the cop-
ing strategies used tended to match the level of appraised
controllability of the stressor—that is, we hypothesized that ap-
praised controllability would be positively related to the use of
problem-focused coping and negatively to the use of emo-
tion-focused coping. We further examined two alternative hy-
potheses regarding the relation of coping and distress: The
main-effects hypothesis, which suggests that problem-focused
coping strategies are generally more effective in reducing dis-
tress than emotion-focused strategies (regardless of control ap-
praisal of the stressor), and the goodness-of-fit hypothesis of
control appraisal and coping, which predicts that type of coping
The empirical literature examining the goodness-of-fit hy-
pothesis is small and almost exclusively cross-sectional.
Forsythe and Compas (10), for example, reported a significant

160
Zakowski et al.
Annals of Behavioral Medicine
(emotion-focused or problem-focused) would interact with con-
trol appraisal such that the combination of low control and high
emotion-focused coping and high control and low emotion-fo-
cused coping would be associated with the least stress, and vice
versa for problem-focused coping.
ested in examining emotion-focused and problem-focused cop-
ing, we combined the self-blame, wishful thinking, and
avoidance subscales into one emotion-focused coping score
(13). The problem-focused coping scale was left intact. Zero-or-
der correlations revealed moderate intercorrelations among the
three emotion-focused subscales and a near zero correlation of
the total emotion-focused coping score and problem-focused
coping (see Table 1). Support seeking was not included in either
the emotion- or problem-focused category as it includes ele-
ments of both types of coping.
Two sets of scores were computed for each coping strategy,
raw scores, and relative percentage scores. Raw scores were
simply calculated by summing the scores for each scale. How-
ever, it has been suggested that these scores may be biased by
differential numbers of items on each scale and may not reflect a
person’s relative coping effort (5). Therefore, all results were
verified using relative scores that were calculated according to
methods proposed by Vitaliano et al. (5).
METHOD
Participants
Participants were recruited through Walter Reed Army
Medical Center and advertisements in the Washington, DC area,
as part of a larger longitudinal study of stress in parents of
healthy and chronically ill children. Participants included 44
women and 28 men who were between 24 and 60 years of age
(M = 36.01, SD = 7.4). Most (93%) were married, 48.6 % had at
least a college degree, 58.9% had an annual household income
of over $30,000, and 78.3% worked outside the home.
First, the mean score was obtained for each subscale by di-
viding the raw score by the number of items on the scale. The
relative score was then calculated by dividing the mean score for
the particular scale by the sum of the mean scores of all of the
scales. This was calculated for the combined emotion-focused
score and the problem-focused subscale. In keeping with pub-
lished methods (5), support seeking was included in the denomi-
nator when calculating percentages of problem and emotion-fo-
cused coping, such that relative percent scores reflect a
percentage of all coping efforts used including support seeking.
Procedures
All study sessions were conducted in the participants’
homes to facilitate study participation. The study was intro-
duced at an initial session (Time 0) during which all procedures
were explained and informed consent was obtained. Subsequent
sessions were scheduled approximately 2 months apart, at
which time questionnaires and tasks were administered. Each
session lasted approximately 45 min. Only data from the first
two assessments (Time 1 and Time 2) were used for this study.
The measures that were included for the analyses are described
later. They were administered in the following order: At Time 1
participants completed the Symptom Checklist-90–Revised
(SCL-90R) followed by the proofreading task; at Time 2 partici-
pants completed the SCL-90R, the proofreading task, Demo-
graphics, and the Ways of Coping Checklist–Revised
(WCCL–R).
SCL-90–R (19). Participants were asked to rate how much
they had been bothered by each of 90 symptoms over the past 2
weeks on a 5-point rating scale ranging from 0 (not at all) to 4
(extremely). This measure yields nine subscales: Somatization,
Obsessive–Compulsive, Interpersonal Sensitivity, Depression,
Anxiety, Hostility, Phobic Anxiety, Paranoid Ideation, and
Psychoticism. Validity and reliability are well established (20).
Only the Depression and Anxiety subscales were of interest for
this study, as these symptoms are thought to be reflective of
overall distress and have been examined in previous studies of
this nature (21). Internal consistency for these two subscales are
.90 and .85, respectively (20). This measure was administered at
Time 1 and Time 2.
Measures
Demographic variables. Basic demographic information
was obtained including age, race, income, education, and gen-
der using a face valid questionnaire.
WCCL–R (18). This revised version of the original 68-item
WCCL (7) included 42 items loading on five factors (prob-
lem-focused, support seeking, self-blame, wishful thinking, and
avoidance coping). This questionnaire was administered at
Time 2. Following the method described by Folkman and Laza-
rus (7), respondents were asked to list a major stressful event or
experience that had occurred in the past year and to refer to that
event when answering the subsequent questions. Participants
then rated how frequently they used each particular coping strat-
egy in dealing with this event on a 4-point scale ranging from 0
(never used) to 3 (regularly used). Reliability of the subscales (α
= .74–.88), and construct validity have been established (18). In
addition, participants were asked the degree to which they felt
the event was controllable on a 5-point scale ranging from 1
(strongly disagree) to 5 (strongly agree). Because we were inter-
TABLE 1
Correlations Among Subscales of the
Ways of Coping Checklist–Revised
1
2
3
4
5
1. Self-blame
2. Wishful thinking
3. Avoidance
4. Support seeking
5. PFC
6. EFC (summary)
.35**
.49**
.00
.49**
.02
.38** –.26*
.64** .84**
–.09
.10
.85** –.03
.30**
–.00
Note. PFC = problem-focused coping; EFC = emotion-focused coping
(summary score of self-blame, wishful thinking, avoidance).
*p < .05. **p < .01.

Volume 23, Number 3, 2001
Coping and Stress
161
Proofreading performance (16). The proofreading task
was included as a behavioral measure of stress. It was derived
from acute stress studies where it has been shown to be a reliable
measure of stressor aftereffects. In addition, performance has
been consistently affected by perceptions of controllability of
the stressor (16,22). Proofreading performance was reduced in
individuals experiencing various life stressors in a number of
studies (e.g., 23,24) and has been found to distinguish between
stress and control groups along with differences in physiologi-
cal and subjective measures of stress (e.g., 17,23,24). Based on
previously published methods (e.g., 16,17), participants were
given a 7-page, double-spaced passage from Jacobs’s The Life
and Death of Great American Cities and were asked to proof-
read the text and circle every error they could find, including ty-
pographical errors, misspellings, and grammatical errors; these
errors had been systematically inserted into the text. Participants
were given 5 min to complete the task. The number of errors
identified was then divided by the number of errors contained in
the passage they read, yielding the percentage of errors accu-
rately identified. It is expected that performance on this task will
decrease as a function of the degree of mismatch between con-
trol appraisals and coping use. Proofreading performance was
assessed at Time 1 and 2 using different passages containing
comparable numbers of errors according to published proce-
dures (e.g., 8,23).
lated stressors were appraised as significantly less controllable
than other types of stressors, F(1, 61) = 8.26, p < .01, confirming
previous findings by Folkman and Lazarus (7). However, use of
emotion-focused and problem-focused coping strategies did not
differ by type of stressor (ps > .05).
The Matching Hypothesis
First, we examined correlations between control appraisals
and use of problem-focused coping and emotion-focused cop-
ing strategies. In accordance with our predictions, perceived
control was negatively correlated with emotion-focused coping
(r = –.33, p < .01), and positively correlated with problem-fo-
cused coping (r = .25, p < .05). Results were similar when rela-
tive coping scores were used (r = –.30, p < .05 and r = .41, p <
.01, respectively).
Main-Effects Versus Goodness-of-Fit
Hypothesis
Analyses next examined predictions of the main-effects and
goodness-of-fit hypotheses, testing relations of coping and dis-
tress. For each of the regression equations the Time 1 distress
measure was entered on Step 1, appraised controllability on Step
2, followed by emotion- or problem-focused coping on Step 3.
Time 1 depression was a significant predictor of Time 2 depres-
sion, predicting 35% of the variance. No other main effects were
found to be significant. Neither appraised control nor coping
were significantly related to the distress measures at Time 2 (ps
> .10). These results were confirmed using relative coping
scores as predictors in Step 3, which showed no significant main
effects on any of the dependent variables (ps > .10).
For each of the regression equations, the interaction term of
Appraised Control × Coping was entered on Step 4 to test the
goodness-of-fit hypothesis. Control and coping scores were first
centered around zero before calculating their cross-product to
minimize multicollinearity (26). The regression lines for the sig-
nificant interactions were then plotted following published meth-
ods (26). The interaction of Control × Emotion-Focused Coping
accounted for significant variance in depression, and proofread-
ing performance, explaining 9.8 and 6.8% of the variance, respec-
tively (see Table 2). For problem-focused coping, no significant
interaction effects emerged. Again, analyses using relative cop-
ing scores confirmed this pattern of results with emotion-focused
coping predicting a significant, albeit slightly smaller proportion
of the variance in Time 2 distress (5.9 and 6.4% for depression
and proofreading performance, respectively). In addition, a sig-
nificant Control × Relative Problem-Focused Coping interaction
emerged only for proofreading performance, predicting 16.2% of
the variance in residualized scores.
RESULTS
Of the 72 participants who completed Time 1 and 2 ques-
tionnaires, 67 had complete data for the proofreading task
(missing values were due to failure to accurately complete the
task). We examined variables for normal distributions and
checked for potential outliers that may skew results. The depres-
sion and anxiety scores from the SCL-90–R were subjected to
logarithmic transformations to reduce skewness to acceptable
levels (25). One participant’s emotion-focused coping score was
discarded because it was over 3 standard deviations above the
mean (25). Demographic variables were examined for their as-
sociations with the main dependent and independent measures.
There was a significant gender difference in proofreading per-
formance at Time 2, with women exhibiting better performance
than men, F(1, 67)= 4.88, p = .03. None of the other demo-
graphic variables were significantly related to any of the main
study measures. Zero-order correlations among dependent mea-
sures revealed a high correlation between the Depression and
Anxiety subscales (rs > .70, p < .001); therefore, only results for
Depression will be reported. All results were comparable when
the Anxiety subscale was entered as a dependent variable.
Sixty-two percent of participants reported health-related
types of stressors, including illness of their child and other
health problems in the family including death. Furthermore,
14% reported employment related stressors, 6% financial stress-
ors, 6% marital problems, and 12% other types of stressors that
did not correspond to any of the aforementioned categories. To
examine any possible differences in perceived control and use of
coping strategies across types of stressors, we compared
health-related stressors to all other types combined. Health-re-
Regression lines were plotted for each of the significant
Control × Coping interactions showing that effects were in the
predicted direction. Figures 1 and 2 depict findings for raw cop-
ing scores. High levels of appraised control in combination with
high levels of emotion-focused coping as well as low control in
combination with low emotion-focused coping were associated
with the highest levels of symptoms of depression and poorest
performance on the proofreading task. Conversely, high con-

162
Zakowski et al.
Annals of Behavioral Medicine
TABLE 2
Regression Analyses of Residualized Scores of Depression and Proofreading Performance on Perceived Control,
Raw Emotion-Focused Coping Scores, and Control × Coping Interactions
R2
∆R2
β
F
df
Depression
Step 1: Time 1 depression
Step 2: Appraised control
Step 3: Emotion-focused coping
Step 4: Control × Coping
Proofreading (% correct)
Step 1: Time 1 proofreading
Step 2: Appraised control
Step 3: Emotion-focused coping
Step 4: Control × Coping
.355
.356
.356
.454
.355
.001
.000
.098
.596
–.035
–.066
.319
39.01***
.13
.39
1, 71
1, 70
1, 69
1, 68
12.21**
.000
.001
.024
.092
.000
.001
.023
.068
–.016
.031
–.161
–.269
.02
.06
1.49
4.62*
1, 65
1, 64
1, 63
1, 62
*p < .05. **p < .01. ***p < .001.
FIGURE 1 Interactions between emotion-focused coping (raw
scores) and appraised controllability on residualized depression scores
of the SCL-90–R (low = –1 SD, high = +1 SD).
FIGURE 2 Interactions between emotion-focused coping (raw
scores) and appraised controllability on residualized proofreading
scores (low = –1 SD, high= +1 SD).
trol/low emotion-focused coping and low control/high emo-
tion-focused coping were associated with the lowest levels of
symptoms and highest proofreading performance (see Figures 1
and 2). The relative problem-focused coping by control interac-
tion showed that high control/high problem-focused coping and
low control/low problem-focused coping were associated with
the highest proofreading performance, an indication of lower
stress levels.
they consider to be relatively uncontrollable, they are more
likely to use coping strategies that regulate their emotions rather
than engage in what may seem to be futile efforts at changing the
situation. Other studies have reported similar associations be-
tween appraisal and use of coping strategies, (e.g., 6,7), and
these findings fit with the theory proposed by Lazarus and
Folkman (1).
The second hypothesis, which proposed that problem-fo-
cused coping would be associated with less distress and emo-
tion-focused coping with more distress regardless of appraised
controllability, was not supported by the data. None of the cop-
ing scales were significant predictors of Time 2 distress after
Time 1 distress was statistically controlled. Previous studies
have found mixed results regarding this main-effects hypothe-
sis. For example, Conway and Terry (11) reported significant
zero-order correlations between coping and depression. How-
DISCUSSION
This study examined hypotheses that have been proposed as
part of Lazarus and Folkman’s (1) transactional model of stress
and coping. As predicted, events that were appraised as rela-
tively uncontrollable were addressed with less problem-focused
and more emotion-focused coping efforts, and vice versa. These
data suggest that when people are confronted with a stressor

Volume 23, Number 3, 2001
Coping and Stress
163
ever, studies that have examined the main effects hypothesis
from both a cross-sectional and a longitudinal perspective (con-
trolling for prior distress symptoms) have reported an attenua-
tion or elimination of the main effects of coping on symptoms
once prior symptom levels were controlled (e.g., 12,27). Exami-
nation of zero-order correlations in this study yielded significant
positive correlations between emotion-focused coping and de-
pression (r = .27, p < .05), suggesting that the main effects re-
ported in the literature may at least in part be confounded by
prior distress that affects both the use of coping strategies and
subsequent distress.
The goodness-of-fit hypothesis, which postulates that a
good match between appraisal and coping strategies will result
in more optimal psychological adjustment, was at least partly
confirmed. On the whole, Appraised Control × Coping interac-
tions were significant for emotion-focused coping. In line with
the hypotheses, low-control appraisals in combination with high
levels of emotion-focused coping and high control combined
with low use of emotion-focused coping was associated with the
least amount of distress. This effect held for self-report and
proofreading performance after prior levels of these measures of
stress were controlled. These findings suggest that emotion-fo-
cused coping may be effective in dealing with stressors that are
judged to be uncontrollable, thus challenging the notion that
emotion-focused coping is an inherently poorer coping mecha-
nism. It is only when dealing with stressors that are appraised as
controllable that greater use of emotion-focused coping appears
to be less effective.
Of interest, Problem-Focused Coping × Control interac-
tions did not predict self-reported distress. This may have been a
result of the relatively small sample size used in this study. In ad-
dition, measurement limitations may provide an explanation for
the null findings on problem-focused coping. The WCCL–R
only includes one problem-focused coping subscale but has
three emotion-focused coping subscales. It is possible that this
scale did not explore all possible problem-focused coping strate-
gies that were important to this population. Replications of the
findings using larger sample sizes and possibly more extensive
measures of problem-focused coping may provide clarity re-
garding the relative importance of problem-focused and emo-
tion-focused coping strategies in moderating the effects of con-
trollable and uncontrollable stressors.
This study is the first to include a behavioral measure of
stress in a test of the goodness-of-fit hypothesis. The
nonsignificant correlation between Time 1 and 2 proofreading
scores should be noted. Performance on this task is situationally
determined, and one would expect correlations to be low, espe-
cially if a stressful event intervened in between the two adminis-
trations, as may be the case for some of the study participants.
Why intercorrelations are lower than for self-report is not clear
and could be due to other influences on performance. Proof-
reading performance has consistently been shown to be affected
by both chronic and acute stress. Results for this task were com-
parable to those obtained using self-report measures, which
strengthens these findings considerably. However, future studies
are needed to replicate these results using behavioral measures
as well as other non-self-report measures of stress (e.g., cardio-
vascular and endocrine measures). Of interest, the proofreading
task revealed an additional finding that was not present for the
self-report measures: a significant relative Problem-Focused
Coping × Control interaction. The reason this finding occurred
only for proofreading and only when using relative scores is not
clear. One may speculate that this interaction suggests that at-
tempts at control (reflected in high problem-focused coping ef-
forts relative to low attempts at regulating emotions) in the face
of an uncontrollable stressor may lead to subsequent helpless
behavior and lack of effective action even on an unrelated and
presumably controllable task. This general helpless behavior
may be reflected in poorer performance on the proofreading
task. Seligman (28) would argue that such learned helplessness
would also be associated with depression; however, low control
and high problem-focused coping were not associated with
higher levels of depression in this study. This interaction was not
found for absolute problem-focused coping scores, which raises
the possibility that such helpless behavior may only ensue if at-
tempts at controlling an uncontrollable situation are not bal-
anced by a comparable effort at regulating distressing emotions.
This interpretation is speculative at best, and future studies may
help to resolve this issue by assessing stress on various dimen-
sions, including psychological, behavioral, and physiological
measures.
A note regarding the two scoring methods of coping—raw
and relative scores—is warranted. The inclusion of both scoring
methods in this study was based on Vitaliano et al.’s (5) argu-
ment that the two methods are conceptually distinct. These au-
thors reasoned that two individuals with identical frequency of
emotion-focused coping (raw scores) may have very different
coping profiles based on how frequently they use other strate-
gies. According to the authors, this would have clinical implica-
tions for their adjustment to stressors. Although this idea has
much intuitive appeal and fits well with Lazarus and Folkman’s
(1) model, few studies have compared the two scoring methods.
Only one other study tested the goodness-of-fit hypothesis using
both scoring methods and found no difference in the results (11).
The importance of this distinction remains to be clarified.
This study is the first to provide support for the good-
ness-of-fit hypothesis in a longitudinal design allowing control
of the possible confound of prior distress levels. Conclusions re-
garding causality, however, should be considered with caution.
One caveat to be noted is the time frame chosen for the coping
measure. Participants were allowed to choose any stressful
event experienced over the past year, thus raising the possibility
that the stressor and associated coping efforts in fact preceded
the “baseline” stress measures in some cases. Thus, although
this study provides some control for prior distress levels, the in-
terpretation is limited by the possibility that this was not a true
baseline for at least part of the study sample. One may argue that
using a 2-month time frame, which would have fallen in be-
tween Time 1 and 2 assessments, may not have completely elim-
inated this problem, as participants may still have chosen to re-
port ongoing stressors the onset of which predated that
prescribed time frame. This issue raises the concern that retro-

164
Zakowski et al.
Annals of Behavioral Medicine
(7) Folkman S, Lazarus RS: An analysis of coping in a middle-aged
community sample. Journal of Health and Social Behavior.
1980, 21:219–239.
(8) Collins DL, Baum A, Singer, JE: Coping with chronic stress at
Three Mile Island: Psychological and biochemical evidence.
Health Psychology. 1983, 2:149–166.
(9) Vaillant GE, Bond M, Vaillant CO: An empirically validated hi-
erarchy of defense mechanisms. Archives of General Psychia-
try. 1986, 43:786–794.
(10) Forsythe CJ, Compas BE: Interaction of cognitive appraisals of
stressful events and coping: Testing the goodness of fit hypothe-
sis. Cognitive Therapy and Research. 1987, 11:473–485.
(11) Conway VJ, Terry DJ: Appraised controllability as a moderator
of the effectiveness of different coping strategies: A test of the
goodness of fit hypothesis. Australian Journal of Psychology.
1992, 44:1–7.
(12) Felton BJ, Revenson TA: Coping with chronic illness: A study
of illness controllability and the influence of coping strategies
on psychological adjustment. Journal of Consulting and Clini-
cal Psychology. 1984, 52:343–353.
(13) Vitaliano PP, DeWolfe DJ, Maiuro RD, Russo J, Katon W: Ap-
praised changeability of a stressor as a modifier of the relation-
ship between coping and depression: A test of the hypothesis of
fit. Journal of Personality and Social Psychology. 1990,
59:582–592.
(14) Baum A, Grunberg N, Singer JE: The use of psychological and
neuroendocrinological measurements in the study of stress.
Health Psychology. 1992, 59:217–236.
(15) Paulhus DL: Measurement and control of response bias. In
Robinson JP, Shaver PR, Wrightsman LS (eds), Measures of
Personality and Social Psychological Attitudes. San Diego,
CA: Academic, 1991, 17–59.
(16) Glass DC, Singer JE: Urban Stress. New York: Academic,
1972.
(17) Baum A, Gatchel RJ, Schaeffer MA: Emotional, behavioral,
and physiological effects of chronic stress at Three Mile Island.
Journal of Consulting and Clinical Psychology. 1983,
51:565–572.
(18) Vitaliano PP, Russo J, Carr JE, Maiur RD, Becker J: The Ways
of Coping Checklist: Revision and psychometric properties.
Multivariate Behavioral Research. 1985, 20:3–26.
(19) Derogatis LR, Cleary PA: Confirmation of the dimensional
structure of the SCL-90: A study in construct validation. Jour-
nal of Clinical Psychology. 1977, 33:981–989.
(20) Derogatis LR: The SCL-90 Manual I: Scoring, Administration,
and Procedures for the SCL-90. Baltimore, MD: Clinical
Psychometric Research, 1977.
(21) Epping-Jordan JE, Compas BE, Osowiecki DM, et al.: Psycho-
logical adjustment in breast cancer: Processes of emotional dis-
tress. Health Psychology. 1999, 18:315–326.
(22) Cohen S: Aftereffects of stress on human performance and so-
cial behavior: A review of research a theory. Psychological Bul-
letin. 1980, 88:82–108.
(23) Gatchel RJ, Schaeffer MA, Baum A: A psychophysiological
field study of stress at Three Mile Island. Psychophysiology.
1985, 22:175–181.
spective reports of stressful events and use of coping strategies is
not the most adequate assessment strategy for this research. A
related issue concerns the question of causality regarding ap-
praisal and coping use. Our study design did not allow us to de-
termine whether the choice of coping strategies followed or pre-
ceded the appraisal process. For example, it is conceivable that
control appraisals would be higher after the person has effec-
tively used problem-focused coping in dealing with a stressor.
This reasoning does not contradict Lazarus and Folkman’s (1)
model, as the authors considered stress and coping to be a con-
tinual feedback process that includes appraisal and reappraisals
of the situation.
Further limitations concern the study sample, including the
relatively small sample size already mentioned and the possibil-
ity that the generalizability of these findings may be limited by
the participant population chosen (approximately 75% of partic-
ipants had a child with a chronic illness). Most previous studies
on this topic have focused on special populations including col-
lege students, psychiatric patients, caregivers, and medical pa-
tients (11–13). Studies using community samples representative
of the general population will increase confidence in the
generalizability of the transactional model.
This research has potential implications for stress manage-
ment interventions and the idea that teaching effective coping
strategies may need to undergo a shift from a model of inher-
ently adaptive and maladaptive strategies (e.g., 9) to a more
complex and flexible one that allows for moderator variables. In
line with Lazarus and Folkman’s theory (1), these results sug-
gest that the person–situation interaction needs to be taken into
account. In fact, studies that have incorporated this model into
broader coping effectiveness training interventions by teaching
individuals to accurately appraise stressor controllability and to
adjust choice of coping strategies accordingly have reported sig-
nificant improvements in coping efficacy, distress, and depres-
sion (29,30). Future research should examine other potential sit-
uational appraisals that may constitute important moderators of
the coping–stress relationship.
REFERENCES
(1) Lazarus RS, Folkman S: Stress Appraisal and Coping. New
York: Springer, 1984.
(2) Folkman S, Lazarus RS: If it changes it must be a process:
Study of emotion and coping during three stages of college ex-
amination. Journal of Personality and Social Psychology. 1985,
48:150–170.
(3) Folkman S, Lazarus RS, Gruen RJ, DeLongis A: Appraisal,
coping, health status, and psychological symptoms. Journal of
Personality and Social Psychology. 1986, 50:571–579.
(4) Moos RH (1974). Psychological techniques in the assessment
of adaptive behavior. In Coelho GV, Hamburg DA, Adams JE
(eds), Coping and Adaptation. New York: Basic Books, 1974,
334–399.
(5) Vitaliano PP, Maiuro RD, Russo J, Becker J: Raw versus rela-
tive scores in the assessment of coping strategies. Journal of Be-
havioral Medicine. 1987, 10:1–18.
(6) Roberts SM: Applicability of the goodness-of-fit hypothesis to
coping with daily hassles. Psychological Reports. 1995,
77:943–954.
(24) Baum A: Stress, intrusive imagery, and chronic distress. Health
Psychology. 1990, 9:653–675.
(25) Tabachnick B, Fidell LS: Using Multivariate Statistics. New
York: HarperCollins, 1996.
(26) Aiken LS, West SG: Multiple Regression: Testing and Inter-
preting Interactions. Newbury Park, CA: Sage, 1991.

Volume 23, Number 3, 2001
Coping and Stress
165
(27) Aldwin CM, Revenson TA: Does coping help? A reexamina-
tion of the relation between coping and mental health. Journal
of Personality and Social Psychology. 1987, 53:337–348.
(28) Seligman MEP: Depression, Helplessness, and Death. San
Francisco: W. H. Freeman, 1975.
(29) Folkman S, Chesney M, McKusick L, et al.: Translating coping
theory into an intervention. In Eckerode J (ed), The Social Con-
text of Coping. New York: Plenum Press, 1991, 239–260.
(30) Chesney M, Folkman S, Chamber D: Coping effectiveness
training for men living with HIV: Preliminary findings. Interna-
tional Journal of STD and AIDS. 1996, 7(Suppl.):75–82.